戴声奎
Personal Homepage
Paper Publications
基于统计结构梯度的行人检测
Hits:

Journal:通信技术

Key Words:】边缘信息; 统计结构梯度; SSG; SVM

Abstract:针对自然背景下的行人检测问题,从人体最显著的轮廓信息出发,提出一种基于统计结构 梯度(SSG, Statistical Structural Gradient)的行人检测算法。采用支持向量机(SVM, Support Vector Machine)算法对数据库中行人样本的SSG 特征进行训练,得到行人检测分类器。提出的SSG 算法在sobel 图像中仅做结构统计和角度幅值计算,能够采用积分图加速行人检测速度。对INRIA 数据库中的288 张样 本进行测试,误检率(FPPW, False Positives Per Window)为10-5,检测率等于90.89%。

Translation or Not:no

Date of Publication:2011-12-07

First Author:DSK

Personal information

Associate professor
Supervisor of Master's Candidates

Gender:Male

Date of Birth:1971-03-25

Education Level:博士研究生

Degree:Doctoral Degree in Engineering

Date of Employment:2007-10-17

Business Address:华侨大学机电实验大楼A618

E-Mail:

Status:在岗

Discipline:Computer Applications Technology
Pattern Recognition and Intelligent Systems
Signal and Information Processing

email :

You are visitors

The Last Update Time : ..


Xiamen Campus: No.668 Jimei Avenue, Xiamen, Fujian, China 361021    Quanzhou Campus: No.269 Chenghua North Rd. Quanzhou, Fujian, China 362021

MOBILE Version